Rust-GPU/rust-gpu

🐉 Making Rust a first-class language and ecosystem for GPU shaders 🚧

What it solves

It replaces the need to write GPU graphics and compute shaders in niche, limited languages like HLSL or GLSL. By bringing Rust to the GPU, it provides developers with a modern language that includes a robust package and module system, built-in safety against memory access errors and race conditions, and high-performance execution.

How it works

The project implements a rustc compiler backend that generates SPIR-V, which is an open compiler target for Vulkan. This allows Rust code to be compiled directly into shaders. The ecosystem includes spirv-std for GPU-specific types and intrinsics, and spirv-builder to integrate GPU crate builds into the CPU build process via build.rs.

Who it’s for

Graphics programmers and game developers who want to use a safe, high-level language for GPU programming instead of traditional shading languages.

Highlights

  • Rust Ecosystem: Enables the use of Cargo and crates.io for developing and publishing GPU crates.
  • Safety: Leverages Rust's safety guarantees to prevent out-of-bounds memory access and race conditions on the GPU.
  • SPIR-V Target: Specifically targets the SPIR-V intermediate representation for compatibility with Vulkan.
  • Integrated Tooling: Provides a dedicated builder to simplify the compilation of GPU code within standard Rust projects.

Related

  • Project
    NVlabs/cuda-oxide

    A custom rustc backend that enables writing and compiling GPU kernels in pure Rust, providing a single-source workflow for host and device code.

  • Project
    Rust-GPU/rust-cuda

    A set of libraries and tools that allow developers to write and execute high-performance GPU code fully in Rust for NVIDIA CUDA hardware.

  • Project
    NVlabs/cutile-rs

    A tile-based system for writing memory-safe, data-race-free GPU kernels in idiomatic Rust, utilizing JIT compilation via CUDA Tile IR.

  • Dispatch
    cuda-oxide: Bringing Rust's Safety and Expressiveness to CUDA

    NVIDIA introduces cuda-oxide, an experimental Rust-to-CUDA compiler that allows developers to write GPU kernels in idiomatic Rust without relying on DSLs or foreign language bindings.

  • Project
    tracel-ai/cubecl

    A Rust language extension and JIT compiler for writing high-performance compute kernels that run across NVIDIA, AMD, Apple, and CPU platforms.